Spatio-temporal learning by the ant ectatomma ruidum

We tested, under field and laboratory conditions, whether the neotropical ant Ectatomma ruidum Roger can learn several associations between temporal and spatial changes in the daily pattern of food availability. Honey was shuffled between two or three feeding sites following a fixed daily schedule. Foragers learnt to associate particular sites with the specific times at which food was available, individually marked ants being observed on the correct sites at the correct times. Some ants anticipated the time of food delivery by approximately 30 min, and it was not necessary for them to be rewarded at the first stage of the sequence of food collection to continue their search for honey according to the correct schedule of reward. Ants also followed the same schedule when no honey was supplied at each stage of the sequence, and they stayed at the expected unrewarded site for a period equivalent to the reward period of the corresponding training phase, indicating that they had learnt when and for how long the food was available. Thus, ants rely on their spatio-temporal memory rather than on local cues coming from the honey source to guide them.